Overriding inlining of rdf:type

[johanwk]

I have an ontology with both classes and individuals. It would be good to not inline some cases of rdf:type, to emphasise how various individuals belong to different parts of the class taxonomy.

Is there a way to request the opposite of puml:InlineProperty? Preferably, this would be applied only to selected rdf:type triples (the default format with inlining of rdf:type helps a lot to avoid busy diagrams).

[VladimirAlexiev]

@johanwk Give me an example, because I think this works now:

• https://github.com/VladimirAlexiev/rdf2rml/tree/master/test/unicode is a delightful example about icecream. Notice how the instance (I) (bottom right) refers to a class that's inlined in the instance, but is also shown as (C) node (top left)
• https://github.com/VladimirAlexiev/rdf2rml/tree/master/test/blank-types shows classes defined with restrictions
• https://github.com/VladimirAlexiev/rdf2rml/tree/master/test/vlaanderen shows lots of classes (it's an ugly example)

[johanwk]

Please take a look at the following example? It uses regular arrows marked with "a" for rdf:type.

ex-artefact-make-ttl.pdf

[VladimirAlexiev]

@johanwk Got it. Here's the source so I can use it as an example:

lis:Person a owl:Class.
lis:Artefact a owl:Class.
ex:InstrumentationTechnician a owl:Class; rdfs:subClassOf lis:Person.
ex:PressureGaugeAssembly a owl:Class; rdfs:subClassOf lis:Artefact.
ex:P-12345-PA-01 a ex:PressureGaugeAssembly.
ex:John a ex:InstrumentationTechnician.
ex:John ex:makes ex:P-12345-PA-01.

####################
owl:Class                    puml:stereotype "(C,orange)".
ex:InstrumentationTechnician puml:stereotype "(P,red)".
ex:PressureGaugeAssembly     puml:stereotype "(A,green)".
rdf:type                     puml:arrow puml:up.
rdfs:subClassOf              puml:arrow puml:tri-up.

• I can look for outgoing links of the class, and not inline if it has such links (eg rdf:type, rdfs:subClassOf).
• https://github.com/VladimirAlexiev/rdf2rml/blob/master/doc/rdfpuml.md#reification shows an option puml:NoReify. Do we also need a similar option puml:NoInline?
• I see a bit of clash on puml:stereotype:
  • when applied on an instance, this stereotype is used for that node
  • when applied on a class, this stereotype is used for all instances of the class
  • but when you start treating classes as instances as well... Should ex:InstrumentationTechnician show (C) because it's a class or (P) because that's the stereotype for its instances? Maybe I can disentangle this...
• Also, rdfs:subClassOf wants the -|> arrow

[VladimirAlexiev]

https://github.com/VladimirAlexiev/rdf2rml/tree/master/test/blank-types#readme